ABSTRACT Aggressive variant prostate cancer (AVPC) arises in men who have failed treatment with second-generation anti-androgen therapy. Not driven by the androgen receptor-signaling axis, effective treatment options do not exist for AVPC and new, innovative therapies are urgently needed. Critical to the development of novel therapeutics is the ability to accurately image disease burden in AVPC patients. An imaging modality that can detect both the bone and visceral metastases associated with AVPC does not exist. The overall goal of this project is to develop positron emission tomography (PET) imaging probes and radioimmunotherapy (RIT) agents by targeting a newly identified cell-surface antigen unique to AVPC with novel antibody constructs. Using human antibody phage display, we recently identified a single chain variable fragment (scFv) that specifically bound to a glycosylation-independent epitope on the peptide backbone of the transmembrane protein CD133. Often characterized as a cancer stem cell marker, the function of CD133 in cancer is unknown. All commercially available antibodies for CD133 recognize glycosylation-dependent epitopes that vary between cells and at different stages of the cell cycle. Immunohistochemistry (IHC) with these antibodies yield inconsistent results and poor staining quality; thus contributing to our limited knowledge of CD133. Our scFv for CD133, termed A10, was converted into a full-length immunoglobulin (IA10) for IHC analysis on patient biopsies and tissue microarrays. Remarkably, we found that CD133 was only expressed in bone and visceral metastases of men who had developed AVPC. As a single-photon emission computed tomography imaging agent, IA10 was able to image CD133 expression in vivo. In biodistribution studies, IA10 and a smaller minibody construct version (MA10) both demonstrated high tumor uptake and favorable pharmacokinetics. The high tumor uptake of the A10 antibody constructs was the direct result of their rapid internalization by CD133-expressing cells making them ideal candidates for longitudinal PET imaging and RIT. In this proposal, we will evaluate the utility of IA10 and MA10 as PET probes in cell line-derived and patient-derived xenograft models of AVPC and evaluate CD133 as a pharmacodynamic biomarker (Aim 1) then determine the therapeutic potential of IA10 radiolabeled with 177Lu for beta particle RIT in subcutaneous and metastatic xenografts using single and fractionated doses (Aim 2). Our preliminary data strongly suggest that we have developed potent tools for AVPC that possess the potential to result in a dramatic shift in how the disease is treated.